Design and exploration of novel boronic acid inhibitors reveals important interactions with a clavulanic acid-resistant sulfhydryl-variable (SHV) β-lactamase

Article abstract of DOI:10.1021/jm301490d

Inhibitor resistant (IR) class A β-lactamases pose a significant threat to many current antibiotic combinations. The K234R substitution in the SHV β-lactamase, from Klebsiella pneumoniae, results in resistance to ampicillin/clavulanate. After site-saturation mutagenesis of Lys-234 in SHV, microbiological and biochemical characterization of the resulting β-lactamases revealed that only -Arg conferred resistance to ampicillin/clavulanate. X-ray crystallography revealed two conformations of Arg-234 and Ser-130 in SHV K234R. The movement of Ser-130 is the principal cause of the observed clavulanate resistance. A panel of boronic acid inhibitors was designed and tested against SHV-1 and SHV K234R. A chiral ampicillin analogue was discovered to have a 2.4 ± 0.2 nM K_i for SHV K234R; the chiral ampicillin analogue formed a more complex hydrogen-bonding network in SHV K234R vs SHV-1. Consideration of the spatial position of Ser-130 and Lys-234 and this hydrogen-bonding network will be important in the design of novel antibiotics targeting IR β-lactamases.

Full text of DOI:10.1021/jm301490d

Article
pubs.acs.org/jmc
Design and Exploration of Novel Boronic Acid Inhibitors Reveals
Important Interactions with a Clavulanic Acid-Resistant Sulfhydryl-
Variable (SHV) β‑Lactamase
Marisa L. Winkler,† Elizabeth A. Rodkey,‡ Magdalena A. Taracila,§ Sarah M. Drawz,∥
Christopher R. Bethel,# Krisztina M. Papp-Wallace,§,# Kerri M. Smith,▽ Yan Xu,▽
Jeffrey R. Dwulit-Smith,‡ Chiara Romagnoli,○ Emilia Caselli,○ Fabio Prati,○ Focco van den Akker,*,‡
and Robert A. Bonomo*,†,§,⊥,#
‡
§
∥
⊥
†Departments of Microbiology and Molecular Biology, Biochemistry, Medicine, Pathology, and Pharmacology, Case Western
Reserve University, Cleveland, Ohio 44106, United States
#Research Service, Louis Stokes Cleveland Department of Veterans Affairs Medical Center, Cleveland, Ohio 44106, United States
▽Department of Chemistry, Cleveland State University, Cleveland Ohio 44115, United States
○Department of Life Sciences, University of Modena and Reggio Emilia, Modena, Italy
S
* Supporting Information
ABSTRACT: Inhibitor resistant (IR) class A β-lactamases pose a significant threat to
many current antibiotic combinations. The K234R substitution in the SHV β-lactamase,
from Klebsiella pneumoniae, results in resistance to ampicillin/clavulanate. After site-
saturation mutagenesis of Lys-234 in SHV, microbiological and biochemical character-
ization of the resulting β-lactamases revealed that only −Arg conferred resistance to
ampicillin/clavulanate. X-ray crystallography revealed two conformations of Arg-234 and
Ser-130 in SHV K234R. The movement of Ser-130 is the principal cause of the observed
clavulanate resistance. A panel of boronic acid inhibitors was designed and tested against
SHV-1 and SHV K234R. A chiral ampicillin analogue was discovered to have a 2.4 0.2
nM Ki for SHV K234R; the chiral ampicillin analogue formed a more complex hydrogen-
bonding network in SHV K234R vs SHV-1. Consideration of the spatial position of Ser-
130 and Lys-234 and this hydrogen-bonding network will be important in the design of
novel antibiotics targeting IR β-lactamases.
β-lactamases. There are three β-lactamase inhibitors currently
on the market: clavulanate, sulbactam, and tazobactam (Figure
1), which are generally effective mostly against class A β-
lactamases.2,9 β-Lactamase inhibitors form stable covalent
complexes with the β-lactamase enzymes, which lead to their
inactivation (Scheme 1).2 Unfortunately, β-lactamases have also
evolved resistance to inhibitors; point mutations in blaTEM and
blaSHV give rise to amino acid substitutions in the parent
enzyme that result in inhibitor resistant (IR) β-lactamases that
manifest as elevated MICs and Kis to β-lactam/β-lactamase
inhibitor combinations.2,9
In the SHV family of β-lactamase enzymes, amino acid
substitutions are found that lead to IR variants.8,10−21 The first
IR SHV variant was detected in E. coli and designated SHV-
10.13 The substitution contributing to clavulanate resistance in
this variant was S130G (Ambler numbering22). S130G in the
SHV β-lactamase has been studied extensively.16,19,21 Sub-
sequently, substitutions at amino acid position Met-69 (SHV-
49, M69I) and amino acid position Thr-235 (SHV-107,
INTRODUCTION
■
Nearly 1.7 million healthcare-associated infections in the
United States each year result in almost 100,000 deaths.1,2 β-
Lactam antibiotics are often given to treat such infections. In
general, β-lactams act by attacking the penicillin-binding
proteins (PBPs) of the bacterial cell wall, leading to disruption
of the peptidoglycan polymer and bacterial lysis.2−6 Unfortu-
nately, bacteria have evolved mechanisms to elude these
antibiotics, one of which is the production of β-lactamase
enzymes (EC 3.5.2.6).
Four classes of β-lactamases exist; three of these classes (class
A, C, D) use serine as the active site nucleophile and one class
comprises metallo-enzymes requiring Zn2+ for activity (class
B).1,2 Class A enzymes are the most prevalent enzymes in
Enterobacteriaceae (e.g., Escherichia coli and Klebsiella pneumo-
niae), where they are typically plasmid-encoded.1,2,7,8 TEM
(Temoneira) and SHV are two of the enzymes most commonly
found in Enterobacteriaceae.1,2,7
One means of combating the activity of β-lactamase enzymes
has been the design of mechanism-based inhibitors, which
subvert the activity of β-lactamases against antibiotics.2,9 The
inhibitors are given with β-lactam antibiotics to protect against
Received: October 19, 2012
Published: December 19, 2012
© 2012 American Chemical Society
1084
dx.doi.org/10.1021/jm301490d | J. Med. Chem. 2013, 56, 1084−1097

Journal of Medicinal Chemistry
Article
Figure 1. Chemical structures of the compounds tested in this paper.
On the basis of this reasoning, position 234 was explored to
determine if additional amino acid substitutions would maintain
the IR phenotype. To achieve this goal, site-saturation and site-
directed mutagenesis were conducted to evaluate all 19 amino
acid substitutions. Additionally, we probed the flexibility of the
active site and enzyme−inhibitor or enzyme−substrate
interactions in the K234R variant with the goal of obtaining
insights that could aid in the design of novel antibiotics and
inhibitors. To achieve this second objective, we synthesized a
series of boronic acid transition state inhibitors (BATSIs). We
were next compelled to determine changes in the structure of
the K234R variant through X-ray crystallography that lead to
the IR phenotype. From these results and previous studies, we
propose a novel mechanism underlying inhibitor resistance
involving Lys-234 and displacement of Ser-130, which prevents
or slows proton transfer and inactivation.
Scheme 1. Scheme Showing the Breakdown of a β-Lactam
(S) by the Enzyme (E) to Form the Inactive β-Lactam (P)
and the Interaction of an Inhibitor (I) with the Enzyme to
Either Form an Inactivated Inhibitor (P) and Active Enzyme
or Inactivation of the Enzyme through a Long-Lived
Inhibitor Complex (E−I*)
T235A) were also discovered that result in amoxicillin/
clavulanate resistance.10,15
RESULTS
■
Three different IR SHV isolates were recently reported
containing a K234R substitution (SHV-56, SHV-72, and SHV-
84).11,12,14 Studies of these clinical isolates reported that
position 234 is a key residue leading to IR in the SHV β-
lactamase. Interestingly, Lys-234 is also part of the conserved
KTG motif in class A β-lactamases.22,23 In SHV, Lys-234 is on
the β3 strand, which forms a wall of the active site cavity.14,23
The proximity of Lys-234 to the active site of SHV and the
frequent appearance of the K234R substitution among IR SHV
enzymes lead us to postulate that this residue is critically
important in the evolution of the IR phenotype in SHV.
Construction of Variants at Ambler Position 234;
Mutagenesis and Immunoblotting. Using degenerate
oligonucleotides, we first performed site-saturation mutagenesis
on the blaSHV‑1 gene at Ambler position 234. Thirteen of the 19
amino acid substitutions were obtained in the initial sequencing
screen (100 blaSHV genes selected from E. coli ElectroMAX
DH10B transformants and sequenced). blaSHV K234L
,
blaSHV K234Q, blaSHV K234G, blaSHV K234C, blaSHV K234Y, and
blaSHV K234F were constructed by site-directed mutagenesis.
Expression of the β-lactamase proteins were confirmed by
1085
dx.doi.org/10.1021/jm301490d | J. Med. Chem. 2013, 56, 1084−1097

Journal of Medicinal Chemistry
Article
immunoblotting (Supporting Information Figure 1). We
observed that in E. coli there was lower expression of
K234W, -Y, -F, -V, and -P substituted proteins.
than wild-type SHV-1 for ampicillin/clavulanic acid (16 vs 4
μg/mL, respectively, Table 1). The K234R MICs for
ampicillin/sulbactam and ampicillin/tazobactam were lower
than that of wild-type SHV-1 but significantly higher than for
any other amino acid substitution at position 234. This
indicates that this is a “restrictive position”, with only the
K234R substitution maintaining inhibitor resistance.24 We have
previously observed SHV enzymes that are resistant to
ampicillin/clavulanate, and susceptible to ampicillin/sulbactam
and ampicillin/tazobactam.17,18
Kinetics of K234R with Representative β-Lactam
Substrates. Both SHV-1 and K234R were purified as
described and their identity was confirmed by mass
spectrometry (Figure 2). Kinetic data collected for the
K234R variant relative to values obtained for the wild-type
SHV-1 enzyme are shown in Table 2. Overall, the kcat/Km
Antimicrobial Susceptibility Testing. To assess the
impact of the single amino acid substitutions at Lys234 on β-
lactam and inhibitor susceptibility, minimum inhibitory
concentrations (MICs) for all 19 variants against a panel of
antibiotics and inhibitor combinations (ampicillin (amp),
piperacillin (pip), cephalothin (thin), ampicillin/clavulanate
(amp/clav), ampicillin/sulbactam (amp/sul), and ampicillin/
tazobactam (amp/tazo) were determined in a uniform E. coli
DH10B background. Results are shown in Table 1.
Table 1. MIC Values (μg/mL) of E. coli DH10B Expressing
a
SHV-1 and Lys-234 Variants
amp/
clav
amp/
sul
amp/
tazo
amp
pip
thin
Table 2. Kinetic Properties of SHV-1 and K234R for Three
Representative β-Lactams
E. coli DH10B
1
1
2
<0.06
4
<0.06
512
64
<0.06
128
16
E. coli blaSHV‑1
>16384
16384
>4096
2048
256
16
SHV-1
K234R
E. coli
blaSHV K234R
16
Piperacillin
E. coli
32
4
8
4
8
8
<0.06
<0.06
<0.06
<0.06
<0.06
<0.06
Km (μM)
kcat (s−1
cat/Km (μM−1 s−1
74.6 11.7
305 47
blaSHV K234A
)
1179 65
15.8 0.6
2736 28
E. coli
blaSHV K234X
b
k
)
9
0.2
a
ΔΔGcat (kcal/mol)
+0.33
Inhibitors were calculated in the presence of 50 μg/mL ampicillin.
Includes other 16 amino acid substitutions.
b
Nitrocefin
24.7
Km (μM)
5
10.4 1.7
45.2 0.5
4.3 0.5
kcat (s−1
cat/Km (μM−1 s−1
)
347.8 21.5
14.1 0.22
k
)
All Lys-234 variants (here represented as K234X) with the
ΔΔGcat (kcal/mol)
+0.70
exception of -R have lower MICs for the penicillins compared
to SHV-1. The ampicillin and piperacillin MICs (16384 and
2048 μg/mL, respectively) for the K234R variant expressed in
E. coli DH10B were preserved compared to the other Lys-234
variants, indicating that the Arg substitution maintains the
ability to hydrolyze penicillins. Surprisingly, the E. coli K234A
variant MICs (32 and 8 μg/mL for ampicillin and piperacillin,
respectively) indicate that an alanine substitution at Ambler
position 234 also allows some hydrolytic activity to be
maintained for the penicillins relatively to the other amino
acid substitutions (≤4 and ≤8 μg/mL for ampicillin and
piperacillin, respectively). The K234A substitution was not
further explored due to the lower level of ampicillin resistance
relative to SHV-1 and K234R. When MICs were performed
against cephalothin we found that all of the Lys-234
substitutions had increased susceptibility relative to wild-type
SHV-1.
Cephalothin
Km (μM)
48.9 11.2
42.6 3.4
0.87 0.3
29
4.6
5
1
kcat (s−1
cat/Km (μM−1 s−1
ΔΔGcat (kcal/mol)
)
k
)
0.16 0.3
+1.00
values were lower for the K234R variant. A higher Km value for
cephalosporins (nitrocefin and cephalothin) but a lower Km for
piperacillin was demonstrated by the K234R variant. Except for
piperacillin, we observed that the K234R variant has a lower kcat
for all substrates tested.
Kinetics of K234R with β-Lactamase Inhibitors. Table 3
summarizes the kinetics for SHV-1 and K234R when
inactivated by various β-lactamase inhibitors. The K234R
variant demonstrated a higher Ki than the wild-type SHV-1 for
all three inhibitors (clavulanate, tazobactam, and sulbactam).
The largest increase was for clavulanic acid. The partition ratios
(kcat/kinact) were also determined for the three suicide
For the β-lactam/β-lactamase inhibitor combinations, only
the K234R substitution in E. coli DH10B had a higher MIC
Figure 2. Deconvoluted ESI-MS spectra of (A) SHV-1 and (B) K234R β-lactamases. All measurements have an error of 3amu.
1086
dx.doi.org/10.1021/jm301490d | J. Med. Chem. 2013, 56, 1084−1097

Journal of Medicinal Chemistry
Article
destabilize proteins, and we were interested in overall structural
changes of K234R that may be leading to the IR phenotype.
The CD spectra and melting curves for SHV-1 and K234R are
shown in Figures 3A and 33B, respectively. The wild-type SHV-
1 required 6.2 °C more for thermal denaturation (a melting
temperature of 54.0 °C vs 48.2 °C for K234R).
Table 3. Kinetic Properties of SHV-1 and K234R for
Clavulanate, Tazobactam, Sulbactam, and BATSI
Compounds
SHV-1
K234R
Clavulanate
Ki (μM)
1.3 0.1
25
7.7 0.7
50
kcat/kinact (tn)
Tazobactam
Ki (μM)
0.44 0.04
50
0.9 0.1
10
kcat/kinact (tn)
Sulbactam
Ki (μM)
1.6 0.2
12,000
4.6 0.5
60
kcat/kinact (tn)
Achiral Cephalothin BATSI
Ki (μM)
20
2
61
6
Chiral Cephalothin BATSI
Ki (μM)
0.46 0.05
0.037 0.004
0.056 0.005
0.012 0.001
0.048 0.005
0.0024 0.0002
0.0135 0.001
Nafcillin BATSI
Ki (μM)
Figure 3. (A) Far UV spectra of SHV-1 (dashed) and K234R (solid
black) measured at 10 μM enzyme concentration. (B) Fraction of
folded protein was calculated as described and is plotted here relative
to the temperature in order to compute the Tm of each enzyme SHV-1
(dashed) has a Tm of 54.2 °C and K234R (solid black) has a melting
temperature of 48.2 °C.
Chiral Ampicillin BATSI
Ki (μM)
Chiral Cefoperazone BATSI
Ki (μM) 0.0093 0.001
inhibitors; the lower value for sulbactam and tazobactam for the
K234R variant relative to SHV-1 is consistent with the
increased susceptibility of the variant to these inhibitors (see
MICs above).
Assessing the Changes in Secondary Structure and
Stability of SHV-1 and K234R in the Presence of
Inhibitors Using CD. Far UV CD was also used to measure
changes in the secondary structure and thermal stability of
SHV-1 and the K234R variant in the presence of different
inhibitors. These results are shown in Figure 4 and Table 4.
From these measurements, only the binding of the chiral
ampicillin boronate affects the melting temperature of either
SHV-1 or K234R by increases of 7.2 °C for SHV-1 and of 6.6
°C for K234R (Figure 4B,D and Table 5). Tazobactam,
sulbactam, and the achiral BATSI affect the secondary structure
of K234R but not SHV-1 (Figure 4A,C). Clavulanate binding
affects the secondary structure of both enzymes.
Crystal Structure of Apo K234R. The K234R enzyme
crystal structure was solved to high resolution, 1.09 Å, with R
and Rfree values of 0.12 and 0.15, respectively (Table 5). The
K234R structure is very similar to that of the wild-type SHV-1
enzyme structure (PDB ID: 1SHV, Figure 5A), as is evident by
an overall Cα root mean square deviation (rmsd) of 0.32 Å.
Ramachandran analysis indicates that 97% of residues reside in
preferred regions, 2% reside in allowed regions, and 1% (two
residues) are outliers. At this resolution, we could confidently
assign alternative conformations for 23 residues. Two of these
residues are in the active site, Ser-130 and Arg-234.
The lysine to arginine substitution at position 234 is
visualized in a 2Fo−Fc map contoured at 1.5σ (Figure 5B)
with the arginine side chain having two conformations, A and B
(occupancies of 65% and 35%, respectively). In the A
conformation, the entire guanidium group is visible in the
electron density, whereas this moiety is less clear in the minor B
conformation, which is likely a result of increased flexibility and
decreased occupancy. Most strikingly, Ser-130 has two
conformations: one in the canonical wild-type SHV-1
orientation (−144°), toward Lys-73, which has an occupancy
of 55%. In the other conformation, the hydroxyl of Ser-130
(−68°) is oriented toward Arg-234 and has an occupancy of
45% (Figures 5B,C). There is also a slight change in the
Different transition state analogues, BATSI compounds, were
next investigated to determine interactions that may be
responsible for the IR phenotype of the K234R substitution
and also to elucidate chemical properties that may be important
for novel inhibitor design (Figure 1). The BATSIs tested
include: (1) an achiral compound with the R1 side chain of
cephalothin, (2) a chiral compound with the R1 side chain of
cephalothin and a R2 meta-carboxyphenyl group, (3) a nafcillin
BATSI with the R1 side chain of nafcillin and a R2 methyl-meta-
carboxyphenyl group, (4) a cefoperazone BATSI with the R1
side chain of cefoperazone and a R2 methyl-meta-carboxyphenyl
group, and (5) a chiral ampicillin BATSI with the R1 side chain
of ampicillin and a R2 methyl-meta-carboxyphenyl group.
The chiral cephalothin BATSI demonstrated a 44-fold lower
Ki for SHV-1 than the achiral BATSI (0.46 0.05 and 20
2
μM, respectively). These observations suggested that the meta-
carboxyphenyl group at the R2 position of the compound is
important for interactions with the enzyme leading to its
inhibition. The meta-carboxyphenyl group at the R2 position of
the compound lowers the Ki of the K234R variant for the
compound even more (61 6 vs 0.012 0.001 μM for the
achiral vs chiral cephalothin BATSI, respectively) for a 5000-
fold difference. The chiral nafcillin, chiral cefoperazone, and
chiral ampicillin BATSI analogues have a Ki in the nM range for
SHV-1 (37 4, 9.3 1, and 56 6 nM, respectively) and for
the K234R variant (48
5, 13.5
1.4, and 2.4
0.2 nM,
respectively). The values are comparable for SHV-1 and for
SHVK234R for the chiral nafcillin and chiral cefoperazone
BATSIs, but the chiral ampicillin BATSI has a 20-fold lower Ki
for K234R relative to SHV-1.
Assessing the Secondary Structure and Stability of
SHV-1 and K234R Using Circular Dichroism (CD). We
explored K234R and SHV-1 through thermal stability measure-
ments and CD because active site substitutions are known to
1087
dx.doi.org/10.1021/jm301490d | J. Med. Chem. 2013, 56, 1084−1097

Journal of Medicinal Chemistry
Article
Figure 4. (A) SHV-1 at 10 μM enzyme concentration in the presence of 50 μM of various inhibitors monitored by CD showing changes in the
secondary structure upon the enzyme binding to the inhibitor. (B) Changes in the thermal denaturation curve of SHV-1 at 10 μM enzyme
concentration in the presence of 50 μM of various inhibitors (the chiral ampicillin BATSI is the only compound to significantly shift the curve). (C)
K234R at 10 μM enzyme concentration in the presence of 50 μM of various inhibitors monitored by CD showing changes in the secondary structure
upon the enzyme binding to the inhibitor (large changes are seen with tazobactam and sulbactam, thin solid line and dashed black line, respectively).
Additionally, clavulanate and the achiral cephalothin BATSI seem to increase the flexibility and movement of the protein (dotted and dashed-dotted
black lines, respectively). (D) Changes in the thermal denaturation cruve of K234R at 10 μM enzyme concentration in the presence of 50 μM of
various inhibitors (the chiral ampicillin BATSI is the only compound to greatly shift the curve). Sulbactam seems to cause an initial destabilization of
the K234R variant, with stability recovered after the first few temperature determinations are made.
Table 4. Melting Temperature of SHV-1 and K234R in the Presence of Various Inhibitors
SHV-1
K234R
a
a
Tm (°C)
ΔΔGu (kcal/mol)
ΔHVH (kcal/mol)
Tm (°C)
ΔΔGu (kcal/mol)
ΔHVH (kcal/mol)
54.2
54.4
54.4
55.0
55.1
61.4
116
149
183
113
160
143
48.2
49.0
92
Tazo
+0.07
+0.07
+0.28
+0.32
+2.56
+0.23
114
Sul
Clav
46.1
48.9
54.8
−0.6
+0.2
68
100
119
achiral cephalothin boronate
chiral Amp boronate
+1.88
a
Determined by ΔΔGu = ΔTm × ΔSapo
position of Ser-70 in K234R relative to SHV-1, which likely
contributes to the change in kinetic behavior and hydrolysis
patterns of the variant. All of the other active site residues
occupy almost identical positions as compared to wild-type
SHV-1 (PDB ID: 1SHV), likewise no secondary structure
changes are observed (Figure 5C). There is a HEPES buffer
fragment in the active site of K234R as has been observed
previously in a structure of SHV-2.25 The oxygens of the
HEPES fragment are within hydrogen bonding distance of
residues Thr-235 and Arg-244 as well as both conformations of
Ser-130 and Arg-234 and the backbone nitrogen of Gly-236.
Active site waters, including the deacylation water near Glu-166
and Asn-170, are conserved between the K234R and wild-type
SHV structures, with the exception of a water near Arg-234.
This water is present in the wild-type structure but is displaced
by the second Arg-234 conformation (Figure 5B, red sphere).
Molecular Modeling/Molecular Dynamics Simulation
(MM/MDS). Molecular docking was performed with SHV-1
(Figure 6A,B) and K234R (Figure 6C,D) with the chiral
ampicillin BATSI to understand the high affinity K234R shows
for this inhibitor. To determine the role of Ser-130 in the
inhibitor binding and to corroborate the X-ray structure
findings, a 4 or 6 ps MDS was performed for SHV-1 and
K234R in complex with the chiral ampicillin BATSI. During
SHV-1 minimization and docking, the torsion angle of Ser-130
changes from −60° to −24° (Supporting Information Figure
2A), which shifts Ser-130 from pointing toward Lys-73 to Lys-
234. After docking, the hydroxyl group of Ser-130 is oriented
1088
dx.doi.org/10.1021/jm301490d | J. Med. Chem. 2013, 56, 1084−1097

Journal of Medicinal Chemistry
Article
was able to reduce susceptibility to ampicillin/clavulanate
(Table 1). As observed by Drawz, et al. in studies on the
N276D variant, the K234R variant retains a high level of
resistance to ampicillin and piperacillin, which is not commonly
observed in typical IR enzymes.8 Additionally, all of the K234
variants were less resistant to ampicillin/tazobactam and
ampicillin/sulbactam than the wild-type SHV-1 in E. coli
DH10B. This same susceptibility pattern was observed earlier
in exploring the Asn-276 variants and in investigating the role
of Arg-244 in SHV.8,17,18
Steady-State Kinetics. Steady-state kinetics assays were
performed to find a biochemical basis for the IR phenotype.
Overall, the K234R variant has impaired β-lactamase activity
when compared with the wild-type SHV-1 β-lactamase (kcat/Km
ratios are 18−57% of wild-type) (Table 2). The Km and kcat/Km
were least affected for piperacillin but most affected for
cephalosporins.
Resistance to clavulanate in K234R is manifested as a 7-to-8-
fold higher Ki of the enzyme for the inhibitor; more
importantly, K234R seems to turn over clavulanate better
than SHV-1 (a kcat/kinact or partition ratio of 50 for K234R)
(Table 3). Notably, the K234R variant also shows an increased
Ki for tazobactam and sulbactam relative to SHV-1, which is not
reflected in the MIC measurements (SHV-1 expressed in E. coli
shows more resistance to these inhibitors than the K234R
variant). However, the lower kcat/kinact for sulbactam and
tazobactam for the K234R variant vs SHV-1 (60 vs 12000 and
10 vs 50, respectively) provide an explanation for the higher
susceptibility of the K234R variant to sulbactam and
tazobactam relative to SHV-1.
Table 5. Data Collection and Refinement Statistics
parameter
Data Collection Statistics
value for K234R
space group
P212121
cell dimensions
a, b, c (Å)
49.6, 55.4, 84.0
90.0, 90.0, 90.0
0.97
α, β, γ (Å)
wavelength (Å)
resolution (Å)2
Rsym
1.09
0.03 (0.18)
44.7 (9.6)
84.3 (82.2)
4.6 (4.2)
I/σI
completeness (%)
redundancy
Refinement Statistics
resolution range (Å)
no. of reflections
23.8−1.08
84309
R
work/Rfree
0.12/0.15
2210/49/229
no. of atoms (protein/ligand/water)
rmsd
bond length (Å)
0.015
1.67
bond angle (deg)
av B-factors (Å2)
protein (main chain/side chain/all)
HEPES/cymal-6
8.56/11.15/9.8
11.65/16.17
19.95
water
Ramachandran plot statistics (%)
preferred regions
allowed regions
97.06
1.96
0.98
outliers
Using BATSIs to Probe SHV-1 and the Novel Variant
K234R. BATSIs were used to probe kinetic relationships
among the enzymes. These compounds replace the β-lactam
with a boronic acid moiety and form reversible covalent bonds
with β-lactamase enzymes, allowing measurement of binding
energies and providing an ability to probe the constraints of the
active site of the enzyme.26,27
toward Lys-234 (Figure 6B). The 4 ps MDS of SHV-1 with the
chiral ampicillin boronate shows a similar flexibility for the Ser-
130 torsion angle. The model of SHV-1 and the chiral
ampicillin BATSI show no interaction with the Lys-73 and Asn-
132 residues. In the preacylation model of SHV-1 (Figure 6A),
the hydrogen bonding pattern shows important hydrogen
bonds forming with Ser-70, Asn-132, Asn-170, and Arg-244 for
a total of four hydrogen bonds.
During the K234R molecular dynamic simulation, the χ1
torsion angle of Ser-130 changes from −171° to −129°
(Supporting Information Figure 2B) and the carboxyl of the
chiral ampicillin BATSI makes/breaks hydrogen bonds with
Arg-234 (Figure 6C,D). The preacylation model of K234R
(Figure 6C) has a more extensive hydrogen bonding pattern
with important hydrogen bonds occurring between the chiral
ampicillin BATSI and amino acids Ser-70, Asn-132, Ser-130,
Arg-234, and Arg-244 for a total of six hydrogen bonds. The
hydrogen bonds with Ser-70, Asn-132, and Arg-244 are
preserved in both models (Figure 6). The lack of interactions
between the chiral ampicillin BATSI and Lys-234 in SHV-1
may be due to the difference in Ser-130 mobility angle in SHV-
1 and K234R.
As has been observed before, the meta-carboxyphenyl group
seems to be important for the SHV β-lactamase to have a high
affinity for the BATSI compounds (the achiral cephalothin
analogue lacking this group has a higher Ki than any BATSI
compounds with the m-carboxyphenyl group).26 The other
notable observations are the 38-fold lower Ki of the chiral
cephalothin BATSI for the K234R variant relative to SHV-1
and the 23-fold lower Ki of the chiral ampicillin BATSI for the
K234R variant relative to SHV-1. This lowers the Ki for this
compound against K234R to 2.4 nM. These changes in binding
affinities can be translated into changes in activation energy
(ΔΔGcat, Table 2). This measures the energy required to
proceed from the free β-lactamase and compound to the
transition state. The largest changes are therefore seen with the
chiral ampicillin and chiral cephalothin BATSIs, which have a
ΔΔGcat of −1.9 and −2.0 kcal/mol, indicating the formation of
the transition state with the K234R variant is thermodynami-
cally favored.
Thermal Stability and CD. SHV-1 and K234R were next
studied using CD (Figure 3). The K234R substitution
destabilizes the enzyme by 6.2 °C (Figure 3B). This large
change in Tm in solution was not expected from a single amino
acid substitution. One possible explanation for this observation
is that the larger side chain of arginine is interfering with other
residues in the protein and resulting in the disorder.
Alternatively, the observed movement of Ser-130 and Arg-
DISCUSSION
■
We performed an in-depth examination of position 234 in SHV
to increase our understanding of the phenotype of this
important residue in IR SHVs and to develop insight into
novel compound design.
The Microbiological and Biochemical Impact of
Amino Acid Substitutions at Ambler Position Lys-234
in SHV. Susceptibility testing of all 19 amino acid variants at
position 234 revealed that only the K234R substituted enzyme
1089
dx.doi.org/10.1021/jm301490d | J. Med. Chem. 2013, 56, 1084−1097

Journal of Medicinal Chemistry
Article
Figure 5. (A) Ribbon cartoon overlay of K234R (gold) and SHV-1 (PDB ID: 1SHV, white). Overall rmsd is 0.14 Å, all Cα superposition performed
using SSM Superpose in COOT. Ribbon width is proportionate to B-factor, i.e., larger ribbon radius indicates higher B-factor. (B) SHVK234R active
site. 2Fo−Fc density contoured at 1.5σ is shown for both R234 orientations. Two S130 orientations are also observed, with χ1 angles of −144 and
−68°. χ1 angle defined as the angle between the N, CA, CB and CA, CB, Oγ planes as in Mendonca et. al.12 The red sphere represents the water
molecule found in SHV-1. This space is filled by the A orientation of Arg-234 in the K234R structure, which leads the water molecule to be absent in
the crystal structure of the K234R protein. (C) Active site overlay of SHV-1 (PDB ID: 1SHV, white) and K234R (gold). The change in position of
K/R234, S130, and S70 is evident in this image.
234 in the crystal structure may result in the destabilization of
this protein. However, in accordance with our X-ray
crystallography data, the secondary structure of both proteins
is similar (Figure 3A). One thing that was observed in the X-ray
crystallography image was the absence of a conserved water
molecule near Arg-234 in the K234R structure that is present
near Lys-234 in the SHV-1 structure (Figure 5B, red sphere).
The absence of this water molecule may be a source of
destabilization due to the lack of interactions with amino acid
234 in the K234R structure that are able to occur in SHV-1.
CD was next performed with SHV-1 and K234R with a
variety of inhibitors including the suicide inhibitors (clav-
ulanate, sulbactam, and tazobactam) and BATSI compounds to
observe changes in thermal denaturation and secondary
structure of the enzymes in the presence of the inhibitors
(Figure 4). For SHV-1, none of the compounds greatly affect
the secondary structure of the enzyme although subtle changes
can be seen in the presence of clavulanate (Figure 4A).
Interestingly, the chiral ampicillin BATSI is able to stabilize the
enzyme, increasing the melting temperature by 6.5 °C (Table 5
and Figure 4B). We suspect that this is a result of the added
active site interactions of the meta-carboxyphenyl of the chiral
ampicillin BATSI (56 6 nM Ki) and not a function of the
boronate specifically because the achiral boronate does not
stabilize SHV-1.
The results seen when CD was performed with K234R and
the various inhibitors are more intriguing (Table 5 and Figure
4C,D). There are changes observed in the secondary structure
of the K234R variant in the presence of some of the substrates
(sulbactam, tazobactam, and the chiral ampicillin BATSI)
(Figure 4C). More subtle changes occur in the presence of
clavulanate and the achiral cephalothin BATSI. Sulbactam,
tazobactam, and the chiral ampicillin BATSI seem to decrease
the α-helical content of K234R. Sulbactam has the greatest
effect on these secondary structure changes. We note these
inhibitors appear to change the structure of K234R with
1090
dx.doi.org/10.1021/jm301490d | J. Med. Chem. 2013, 56, 1084−1097

Journal of Medicinal Chemistry
Article
Figure 6. Molecular docking of the chiral ampicillin BATSI (light blue) and SHV-1 in the preacylation state (before dative bond formation) (A) and
after dative bond formation (B) and K234R before dative bond formation (C) and after dative bond formation (D).
binding but do not change the structure of SHV-1. The lower
stability of K234R may allow these compounds to more easily
change the secondary structure of the enzyme.
the hydroxyl group toward Lys-73, whereas the K234R Ser-130
torsion angle is −61 12°, which orients the hydroxyl group
toward Arg-234.
One of the most unanticipated results is the melting curve of
the K234R variant in the presence of sulbactam (Figure 4D).
Sulbactam changes the melting curve of K234R so that it is no
longer a single transition from folded to unfolded. Con-
sequently, a melting point cannot be determined from this
curve because the protein does not undergo a two-state
transition. None of the other compounds show this effect with
the K234R enzyme, and cause this type of change in the
melting curve of SHV-1. Similar results are seen in the melting
temperature measurements of the K234R variant in the
presence of the different inhibitors relative to what was seen
with SHV-1. Only the chiral ampicillin BATSI greatly increases
the melting temperature of the K234R variant (6.6 °C). Again
this seems to reflect the added active site interactions that
resulted in the nanomolar Ki of this compound for the enzyme.
Overall, from the kinetics and CD studies, the chiral
ampicillin boronate seems to be the most intriguing compound
for combating IR variants of SHV with the K234R substitution.
This compound shows a low Ki for the enzyme and seems to
stabilize the β-lactamase as evident from the observed shifts in
melting temperature.
Effect of the K234R Substitution on the Structure of
SHV. Table 5 displays the refinement statistics obtained from
the crystal of K234R. Overall, the structure of K234R is highly
conserved with respect to wild-type SHV-1 (PDB ID: 1SHV)
(Figure 5A,C). The variant active site bears striking
resemblance to that of SHV-1. As mentioned earlier, the
K234R substitution confers inhibitor resistance through
changes in Ser-130.12 In that study, MDS suggested preferred
χ1 torsion angles for Ser-130 in SHV-1 and in K234R.12 For
SHV-1, the final torsion angle was −145 11°, which orients
In this reported structure, we observe both of these
orientations (−68° and −144°, respectively) in a 65/35
occupancy ratio (Figure 5B). We also see two orientations of
Arg-234 (A and B); this likely necessitates the two Ser-130
orientations because in the −144° Ser-130 orientation, the “B”
Arg-234 orientation is too close to Ser-130 (1.76 Å). However,
the “A” Arg-234 orientation is at a reasonable hydrogen
bonding distance to Ser-130 (3.14 Å). This is in contrast to the
wild-type enzyme structure where a single orientation of the
shorter lysine side chain at position 234 would allow for both
Ser-130 orientations (distances from Ser-130 to Lys-234 are
2.94 and 2.65 Å, respectively). Unlike in the K234R enzyme,
when the carboxylic acid group of an inhibitor binds the SHV-1
active site, the Ser-130 residue is free to rotate to an optimal
position for inhibitor binding. We see instability in both Ser-
130 and Arg-234 in K234R; both of these residues participate in
inhibitor binding as they are part of, or are in close proximity
to, the carboxyl binding pocket. Small shifts are also observed
for residues Arg-244 and Glu-166, but those could be due to
the lower resolution of the SHV-1 structure.
On the basis of these structural observations, we suggest that
inhibitor binding is affected, and thereby weakened, by this now
mobile region of the active site. We do observe some local
fluctuations in the Arg-234 and Ser-130 side chains, which may
lead to the instability observed by CD. The lower catalytic
efficiencies (Table 2) confirm the CD observations that there
must be structural changes occurring in K234R. From the
crystal structure, we would expect similar catalytic efficiencies
because the active site configuration is relatively conserved
between SHV-1 and K234R, but we are recording a significant
loss in catalytic efficiency (2−5-fold loss). This supports our
1091
dx.doi.org/10.1021/jm301490d | J. Med. Chem. 2013, 56, 1084−1097

Journal of Medicinal Chemistry
Article
hypothesis that the crystal structure has captured a “snapshot”
of the K234R variant, but other conformations of the β-
lactamase exist in solution, leading to the change in catalytic
properties by kinetic analysis and thermal denaturation by CD.
Molecular Modeling. As a result of the nanomolar affinity
and stabilization of the protein by the chiral ampicillin BATSI,
we performed molecular docking and MDS of SHV-1 and
K234R with this compound. In accordance with our other data,
the MDS revealed a greater flexibility and movement in K234R
compared to SHV-1 (Supporting Information Figure 2A vs B).
Before docking, the X-ray crystal structure of SHV-1 shows Ser-
70 (χ = −114°) toward and hydrogen bonded with Lys-73. In
the preacylation docking comparison of SHV-1 and K234R
with the chiral ampicillin BATSI (Figure 6A vs C), a larger
number of hydrogen bonds are seen between K234R and the
chiral ampicillin BATSI compared to SHV-1 with this
compound (Figure 6C vs A). Hydrogen bonds are seen with
the chiral ampicillin BATSI and Ser-70, Ser-130, Asn-132, Arg-
244, and Arg-234 in K234R (the hydrogen bonds with Lys/
Arg-234 and Arg-244 are absent in SHV-1). The torsion angle
of Ser-130 prevents the interaction of Lys-234 with the chiral
ampicillin BATSI in SHV-1; interactions with Arg-244 are also
not observed (Figure 6A). Additionally, the Ser-70 (χ1 =
−125°) hydroxyl is positioned toward Ser-130 (which points
toward Lys-234). Notably, the Ser-70 angle is different upon
docking of the chiral ampicillin BATSI in K234R (χ1 = −100°,
Figure 6C).
Figure 7. The top figure shows the wild-type enzyme with Ser-130 at a
χ angle of −144° where proton shuffling can occur to allow breakdown
of ampicillin by the β-lactamase. The bottom figure shows the
movement of the hydroxyl group of Ser-130 away from Lys-73 when
Ser-130 is at χ = −68°, which prevents the proton shuffling and
ampicillin breakdown.
The preacylation conformation of K234R with the chiral
ampicillin BATSI compound may allow the dative bond to
form more easily, explaining the lower Ki of the chiral ampicillin
boronate for K234R. This molecular modeling provides us with
information that can be used to design drugs active against
these IR enzymes. Lys-73 and Asn-132 may be important target
sites for interactions between novel compounds and IR
enzymes because two hydrogen bonds are present between
the chiral ampicillin BATSI and these amino acids in K234R
but not in SHV-1. Instead of focusing on interactions with Ser-
70 and Ser-130 as sulbactam, tazobactam, and clavulanate do, it
is important to utilize interactions with other important
residues in order to combat enzymes that have already evolved
inhibitor resistance.
Movement of Ser-130 Explains Variant Behavior.
Quantum mechanical and molecular studies (QM and MM)
have been performed with the TEM class A β-lactamase to
understand the amino acids involved in the β-lactam break-
down reaction.28 These studies indicated that in TEM Ser-130
is involved in the tetrahedral collapse required for deacylation
of the β-lactam substrate to occur. Ser-130 is involved in a
proton relay, donating its proton to the β-lactam thiazolidine
nitrogen while accepting a proton from Lys-73 (Figure 7). In
K234R, one of the variations of Ser-130 moves the hydroxyl
group from the −144° torsion angle to the −68° angle, which
shifts the hydroxyl group away from Lys-73. In the −144°
conformation of Ser-130, the oxygen atom of the hydroxyl
group is 2.2 Å away from the Lys-73 hydrogen. However, in the
−68° conformation, the oxygen is 4.2 Å away from the Lys-73
hydrogen. This greater distance likely prevents the observed
proton shuttle from occurring between Lys-73, Ser-130, and the
β-lactam. The inability of the proton shuttle to occur in one
conformation of Ser-130 in K234R slows the catalytic
breakdown of β-lactams by K234R and could underlie the
kinetic results observed, i.e., the lower kcat and lower kcat/Km for
the tested substrates.
Additionally, Ser-130 is moved away from Ser-70 in the
K234R structure relative to that of SHV-1. In the inactivation
mechanism of class A β-lactamase enzymes by clavulanate,
there is a cross-linking step between Ser-70, clavulanate, and
Ser-130.2 The greater distance between Ser-70 and Ser-130 in
K234R likely prevents this cross-linking from occurring in this
enzyme. This explains the decreased resistance to clavulanate
observed with this enzyme.
Mechanistic Implication. Our analysis of position 234 in
SHV allows us to elucidate a novel pathway among IR variants
of this enzyme involving the movement of Ser-130 and its
relationship to K234R. Ser-130, like Lys-234, is involved in a
conserved loop in class A β-lactamases, the SDN loop, which
also includes Asp-131 and Asn-132.21 When a S130G
substitution is present, resistance to clavulanate is observed.13,21
Ser-130 is involved in the permanent inactivation step of the
inhibitor reaction, where cross-linking occurs between Ser-70,
the inhibitor, and Ser-130.2 A glycine at position 130 prevents
this cross-linking from occurring because the glycine does not
have the hydroxyl group of serine, which allows cross-linking.
Other IR class A β-lactamase variants have been explored
through crystal structures including TEM with a R244S
substitution or M69I/V substitutions, which provide further
information about the importance of Ser-130 in inhibitor
resistance.29 These substitutions have also been observed in IR
variants of SHV.15,17,18 In the crystal structures of the TEM
Met-69 variants, Ser-130 was rotated, which is observed to
diminish the ability of this amino acid to act as a site of cross-
linking in inhibitor inactivation.29 In the R244S structure
(TEM-30), there is a loss of a conserved water molecule which
anchors the C3 carboxylate of the inhibitor.29 This affects the
ability of Ser-130 to react with the inhibitor because it swings
away from this amino acid.29 Thus, in the three previously
studied IR class A β-lactamases (M69I/V, S130G, R244S), the
1092
dx.doi.org/10.1021/jm301490d | J. Med. Chem. 2013, 56, 1084−1097

Journal of Medicinal Chemistry
Article
β-Lactamase Expression and Purification. For kinetic assays
and far-UV circular dichroism (CD), β-lactamase enzymes were
expressed and purified as previously described35 using the pBC SK(−)
plasmid or were expressed and purified from the pGEX-6P-2 plasmid
(GE Healthcare, Piscataway, NJ). For SHV-1, E. coli DH10B cells
containing the blaSHV‑1 gene in pBC SK(−) were grown overnight in
superoptimal broth (SOB) medium, harvested by centrifugation at 4
°C, and frozen at −20 °C. After thawing, the β-lactamase was liberated
using stringent periplasmic fractionation with 40 μg/mL lysozyme
(Sigma Aldrich) and 100 μL of 50 mM EDTA, pH 7.8. Preparative
isoelectric focusing was performed with the lysate in a Sephadex
granulated gel (GE Healthcare, Piscataway, NJ) using ampholines in
the pH range 3.5−10 (Bio-Rad) and running the gel overnight at a
constant power of 8 W on a Multiphor II isoelectric focusing apparatus
(GE Healthcare). Preparative isoelectric focusing fractions were
common cause for the inhibitor resistance is a change in Ser-
130.
Our work on the K234R variant indicates that the K234R
substitution also leads to inhibitor resistance through a
movement of Ser-130, which affects cross-linking with the
inhibitor. A second hypothesis is that the presence of an
arginine instead of a lysine at position 234 affects the
electrostatic environment, which makes it easier for the
breakdown of clavulanate and regeneration of the active
enzyme (a higher partition ratio, kcat/kinact, is seen). Lastly,
the altered interrelationships in the carboxyl binding pocket (as
defined by Ser-130, Lys-234, Thr-235, and Arg-244) may be an
additional cause for the inhibitor resistance and result in the
higher Ki (Figure 5C).30 We have demonstrated that Ser-130 is
the important residue underlying IR class A β-lactamases with a
variety of different amino acid substitutions, (M69I/V, S130G,
R244S, and now K234R) convening to lead to inhibitor
resistance through an effect on the position of Ser-130.
̈
concentrated and purified using AKTA fast protein liquid chromatog-
raphy (GE Healthcare, Piscataway, NJ). For SHVK234R, E. coli Origami
2(DE3) cells containing the blaSHV K234R gene in pGEX-6-P-2 were
grown to OD600 = 0.6 in SOB medium at 37 °C, then 0.1 mM
isopropyl β-D-1-thiogalactopyranoside (IPTG) was added and the
cultures were moved to a shaker at 16 °C overnight. The cells were
harvested by centrifugation at 4 °C and frozen at −20 °C. After
thawing, the β-lactamase was liberated by fractionation with 40 μg/mL
lysozyme (Sigma Aldrich). The lysate was purified on the GSTrap FF
HiTrap affinity column according to the manufacturer’s instructions
(GE Healthcare, Piscataway, NJ). Residual TEM β-lactamase from the
EXPERIMENTAL METHODS
■
Mutagenesis, Sequencing, and Cloning. Using the template
blaSHV‑1 gene in the phagemid vector pBC SK(−) (Stratagene, La Jolla,
CA), we used the Stratagene QuikChange Mutagenesis kit with
degenerate oligonucleotides at Ambler position 234 to make the full
complement of amino acid substitutions. After PCR mutagenesis, we
electroporated the resulting plasmids into E. coli ElectroMAX DH10B
cells (Invitrogen, Carlsbad, CA). These plasmids were isolated, and
DNA sequencing of the blaSHV genes was performed. Variants not
acquired after sequencing of 100 clones were created with specific
mutagenic oligonucleotides that were designed as previously
described.31
̈
pGEX-6P-2 vector was removed using AKTA fast protein liquid
chromatography (FPLC). The GST tag was cleaved using PreScission
Protease (GE Healthcare, Piscataway, NJ). Then the protein was
purified a second time with the GSTrap FF HiTrap affinity column to
separate the SHVK234R and the cleaved GST tag. Purity was assessed by
gel electrophoresis using a 5% stacking, 12% resolving SDS-
polyacrylamide gelelectrophoresis (SDS-PAGE) and was found to be
greater than 95%. Purity was confirmed with mass-spectrometry. β-
Lactamase concentration was determined using a spectrophotometric
assay (Biorad Laboratories, Hercules, CA) with ε = 33585 M−1 cm−1
for SHV-1 and ε = 32095 M−1 cm−1 for the K234R variant.
To aid in purification, the blaSHV K234R gene was cloned into the
pGEX-6P-2 plasmid and transformed into E. coli OrigamiTM 2(DE3)
competent cells (EMD Millipore, Billerica, MA). PCR was performed
to obtain the blaSHV K234R gene from blaSHV K234R. Then restriction
digests were performed on the plasmid and on the cloning vector.
Ligation was performed and colonies were selected on ampicillin and
sequenced.
For crystallization, a different protein expression and purification
procedure was followed. The K234R substitution was introduced into
SHV-1 subcloned into the pET24a+ vector (Novagen, Madison, WI)
without the leader sequence. Mutagenesis was performed using the
QuikChange Lightning Site-Directed Mutagenesis kit (Stratagene, La
Jolla, CA) to change lysine (AAG) to arginine (CGC). Mutagenized
DNA was confirmed by sequencing and transformed into OneShot
BL21(DE3) Star chemically competent E. coli cells (Invitrogen,
Carlsbad, CA), per manufacturer protocol, and plated overnight at 37
°C on lysogeny broth (LB) agar containing 50 μg/mL kanamycin.
Then 100 mL of LB media supplemented with 50 μg/mL kanamycin
was inoculated with the plated colonies. This culture was incubated
overnight at 37 °C with 225 rpm shaking. This was used to inoculate
10 L of LB with 50 μg/mL of kanamycin, which was incubated until
OD600 = 0.5−0.6 was reached. A final concentration of 0.5 mM of
IPTG was added to induce protein expression. Then 3−4 h after
induction, cells were harvested by centrifugation and lysed by
microfluidization on a Microfluidics M-110Y microfluidizer processor
using two passes at 90 psi. Cleared lysate was purified by pIEF as
described above. Nitrocefin positive fractions were pooled, concen-
trated, and loaded on a QSepharose FF (GE Lifesciences, Piscataway,
NJ) anion exchange column using a starting buffer consisting of 20
mM bis-Tris pH 6.5 and using an elution buffer consisting of 20 mM
bis-Tris pH 6.5 and 1 M KCl. The K234R β-lactamase was present in
the flow-through fractions of this column with the elution buffer,
whereas the majority of the impurities were not. Nitrocefin positive
peaks were pooled, concentrated, and loaded onto a Superdex75 gel
filtration column. Protein purity was assessed by SDS-PAGE gel and
was found to be greater than 95%. The protein was concentrated to 5
mg/mL using Vivaspin20 centrifugal concentrator (Sartorious Stedim,
Goettingen, Germany).
Immunoblotting. E. coli. DH10B cells containing the blaSHV K234X
plasmids were grown to OD600 = 0.8 and frozen at −20 °C for 48 h. A
periplasmic extraction of the β-lactamases was performed by addition
of 50 μL of 50 mM Tris-HCl (USB, Cleveland, OH) containing 5 μL
of 40 μg/mL lysozyme (Sigma Aldrich), 5 μL of 10 mM MgSO4
(Teknova, Hollister, CA), and 1 μL of a 1000× dilution of benzonase
(Novagen, Madison, WI), and tubes were allowed to stir for 25 min.
Lastly, 1 μL of 50 mM EDTA was added to each tube followed by
stirring for 5 min. Subsequently, debris was pelleted and 10 μL of
sodium dodecyl sulfate (SDS) loading dye buffer was used to
resuspend the pellet prior to incubation at 100 °C for 10 min.
Immunoblots were performed with an anti-SHV polyclonal antibody
to confirm periplasmic production of the SHV β-lactamases from all 19
constructs as previously described.32 Anti-DnaK (Assay Designs, Ann
Arbor, MI) was used as a loading control as previously described.33
Antimicrobial Susceptibility Testing. Minimum inhibitory
concentrations (MICs) for E. coli DH10B cells expressing the blaSHV-1
or mutant blaSHV Lys234X genes cloned in phagemid pBC SK(−) were
determined by Mueller−Hinton (MH) agar dilution according to the
Clinical and Laboratory Standards Institute (CLSI) guidelines.34 The
MICs for various antibiotics were measured using a Steers replicator
that delivered 10 μL of a diluted overnight culture containing 104
colony forming units. Ampicillin, piperacillin, and cephalothin were
purchased from Sigma (St. Louis, MO). Lithium clavulanate was
purchased from USP (Rockville, MD). Tazobactam was purchased
from Chem-Impex International (Wood Dale, IL). Sulbactam (Pfizer,
La Jolla, CA) was a kind gift. Tazobactam, sulbactam, and clavulanate
were tested in combination with 50 μg/mL ampicillin.
Kinetic Measurements. Steady-state kinetic measurements were
performed using continuous assays at room temperature in an Agilent
1093
dx.doi.org/10.1021/jm301490d | J. Med. Chem. 2013, 56, 1084−1097

Journal of Medicinal Chemistry
Article
Scheme 2. Synthesis Scheme Leading to the Creation of the Various BATSI Compounds (a = Chiral Cephalothin, b = Chiral
a
Nafcillin, c = Chiral Ampicillin, d = Chiral Cefoperazone)
a
Synthesis provided in Supporting Information.
8453 diode array spectrophotometer (Agilent, Palo Alto, CA). Each
assay was performed in 10 mM phosphate-buffered saline (PBS) at pH
7.4. Measurements were obtained using nitrocefin (NCF) (BD
Biosciences, San Jose, CA) (Δε482 = 17400 M−1 cm−1), piperacillin
(Δε235 = −820 M−1 cm−1), and cephalothin (Δε262 = −7660 M−1
cm−1). To investigate the role of position 234 in recognition of
substrates and with changes in the binding pocket in a Lys to Arg
substitution, we used a variety of BATSI compounds (achiral
cephalothin,18 chiral cephalothin (synthesis described here), chiral
nafcillin (synthesis described here), chiral cefoperazone (synthesis
described here), and chiral ampicillin (synthesis described here)).
The kinetic parameters Vmax and Km were obtained with nonlinear
least-squares fit of the data (Henri−Michaelis−Menten equation)
using EnzFitter (Biosoft Corporation, Ferguson, MO) as described in
eq 1.
(Cleveland State University). First, 10 μg of enzyme was desalted and
concentrated for ESI-MS using a C18 ZipTip (Millipore, Billerica,
Massachusetts) according to the manufacturer’s protocol. Then eluted
protein samples were diluted with 50% acetonitrile and 0.2% formic
acid to a concentration of 10 mM, infused at a rate of 0.5 μL per min,
and data were collected for 2 min. Spectra were deconvoluted using
the Applied Biosystems Analyst program.
Circular Dichroism (CD). CD experiments were performed in a
Jasco J-815 spectrometer (Easton, MD) with a Peltier effect
temperature controller (GE Healthcare, Piscataway, NJ). Quartz cells
with a 0.1 cm path length (Hellma, New York) were used for all
experiments. Spectra were obtained with a protein concentration of 10
μM. Compounds were tested at a concentration of 50 μM to ensure
that they did not interfere with the refraction of the light by the
protein in the far UV spectrum. Thermal melting was performed
between 22 and 72 °C with a heating rate of 2 °C/min. Raw
equilibrium denaturation data, monitored by far-UV CD at 208, 214,
and 220 nm, were normalized to the fraction of denatured protein
(fu). The data at 208 nm was used for calculations of melting
temperature (Tm), van’t Hoff enthalpy (ΔHVH), the entropy of
unfolding (ΔSu), and the change in Gibbs free energy of unfolding
(ΔΔGu). The Van’t Hoff plot (ln Keq vs 1/T) was used to calculate
ΔHVH as the slope times −R where R is the ideal gas constant. The y-
intercept is ΔS × R. Equations 4 and 5 were used to determine Tm and
v = (Vmax × [S])/(Km + [S])
(1)
The dissociation constants of the preacylation complex, Kis, were
determined by direct competition assays. In our experiments, we use
the term Ki to represent the Km of the inhibitor. The initial velocity
was measured in the presence of a constant concentration of enzyme
with increasing concentrations of inhibitor against the indicator
substrate, NCF. Care was taken to initiate the reaction with the
addition of both NCF and the inhibitor and to limit the initial velocity
determination to the first 5 s.
ΔHVH
.
Ki values for the BATSIs were determined as described above.
However, because of an incubation effect, SHV β-lactamase and the
boronic acid compounds were preincubated for 5 min in PBS before
initiating the reaction with the addition of substrate (NCF) as
previously described.18,36
ΔG = −RT ln Keq = ΔH − TΔS
ln Keq = 1/T(−ΔH/R) + (ΔS/R)
(4)
(5)
With the assumption of a reversible two-state transition, equilibrium
The Ki data was corrected to account for the affinity of NCF for
constants (Keq) at any given temperature were calculated from eq 6.40
SHV-1 and K234R according to eq 2.37
Keq = fu /1 − fn
(6)
Ki(corrected) = Ki(observed)/[1 + ([S]/KmNCF)]
(2)
The ΔΔGu between SHV-1 and SHVK234R was calculated using the
For the inhibitors, Ki was used in place of Km to obtain the energies
needed to form the preacylation complex with either enzyme.
The activation energy required for bond breaking and formation
(ΔΔGcat) was calculated using the kinetic constants and eq 3 as
previously described.38,39
method of Schellman shown in eq 7.41,42
ΔΔGu = ΔTm × ΔSWT
(7)
In the calculation of ΔΔGu between the apo-enzymes and the
complexes, ΔSapo was substituted for ΔSWT in eq 7.
Molecular Modeling and Molecular Dynamics Simulation
(MM/MDS). The atomic coordinates of SHV-1 (PDB ID: 1SHV) and
the crystal structure reported here were used to construct the acyl−
enzyme complexes. Acyl−enzyme models of the chiral ampicillin
BATSI with the wild-type SHV-1 β-lactamase and K234R were
ΔΔGcat = −RTln((kcat/Kmvariant)/(kcat/KmSHV − 1))
(3)
Mass Spectrometry. Electrospray ionization (ESI)-MS was
performed on SHV-1 and the SHVK234R variant with an Applied
Biosystems (Foster City, California) Q-STAR Elite quadrupole-time-
of-flight mass spectrometer equipped with a TurboIon spray source
1094
dx.doi.org/10.1021/jm301490d | J. Med. Chem. 2013, 56, 1084−1097

Journal of Medicinal Chemistry
Article
constructed as previously described using Discovery Studio 3.1
(Accelrys, Inc. San Diego, CA) molecular modeling software.43 The
K234R variant and the WT molecules were solvated (in a box of water
with 7 Å distance from the molecule to the box) and minimized using
the conjugated gradient method. The chiral ampicillin BATSI structure
was constructed using Fragment Builder tools and was minimized
using a Standard Dynamics Cascade protocol of DS 3.1. The inhibitor
was automatically docked into the active site of the SHV-1 and the
K234R β-lactamase using the CDOCKER module of DS 3.1.44 This
protocol used a CHARMm-based molecular dynamics (MD) scheme
to dock ligands into a receptor binding site. Random conformations
were generated using high-temperature MD. The conformations were
then translated into the binding site. Candidate poses were then
created using random rigid-body rotations followed by simulated
annealing. A final minimization was used to refine the ligand poses.
The resulting conformations were analyzed with the most favorable
positioning of the chiral ampicillin BATSI chosen and the complexes
between the enzyme and inhibitor created as previously described.45
The complex was energy minimized and 4−6 ps molecular dynamic
simulation was performed as previously described.38 The MDS was
performed in three steps: heating/cooling, equilibration, and
production. The target temperature was 300 K, and the long-range
electrostatics were treated with Particle Mesh Ewald. The pressure was
kept constant (NPT) during the equilibration and production steps to
better mimic the experimental conditions for explicitly solvated
systems with periodic boundary conditions (PBC). The distance cutoff
value used for counting nonbonded interaction pairs was 14 Å. During
the MDS the time steps were 0.001 ps, and the atom velocities and
positions were calculated at time points that differ by a half time
step.46−48
then mixed in a 1:1 ratio with well solution, yielding a final drop size of
5 μL. Crystals grew to full size within one week. Data was collected at
Stanford Synchrotron Radiation Lightsource (SSRL), beamline BL9−
2. Images were integrated and scaled using HKL2000.50 K234R
crystallizes in the P212121 space group (Table 1 for data collection and
refinement statistics). The structure was solved using isomorphous
replacement with an existing apo wild-type SHV-1 structure (PDB ID:
1SHV). Refmac5 in the CCP4 suite51 was used to perform restrained
anisotropic refinement and COOT52 was used for model building and
manual refinement. Density for a partial HEPES buffer molecule was
observed in the active site after initial refinement, and the fragment
was included in subsequent refinement steps. Library files for the
HEPES fragment were generated using PRODRG2.53 The coordinates
for the apo K234R structure were deposited in the Protein Data Bank
(PDB ID: 4FCF).
ASSOCIATED CONTENT
* Supporting Information
■
S
Full synthesis of the boronate compounds, Western blot of the
SHV 234 variants, and MD simulation of Ser-130 movement in
SHV-1 and K234R. This material is available free of charge via
the Internet at http://pubs.acs.org.
Accession Codes
The crystal structure can be found as 4FCF in the Protein Data
Bank. The coordinates have been deposited.
AUTHOR INFORMATION
Corresponding Author
■
BATSI Synthesis and Analysis. (Scheme 2) All reactions were
performed under argon using oven-dried glassware and dry solvents.
Anhydrous tetrahydrofuran (THF) and diethyl ether were obtained by
standard methods and freshly distilled over sodium benzophenone
ketyl prior to use under argon. All reagents were purchased from
Sigma-Aldrich and Fluka. Reactions were monitored by thin layer
chromatography (TLC), which were visualized by UV fluorescence
and by Hanessian’s cerium molybdate stain. Chromatographic
purification of the compounds was performed on silica gel (particle
*For F.v.d.A.: E-mail, fxv5@case.edu. For R.A.B.: phone, (216)
791-3800 ext 4399; E-mail, robert.bonomo@med.va.gov;
address, Robert A. Bonomo, MD, 10701 East Boulevard,
Cleveland, Ohio, 44106.
Author Contributions
The manuscript was written through contributions of all
authors. All authors have given approval to the final version of
the manuscript. M.L.W. and E.A.R. contributed equally.
size 0.05−0.20 mm). Melting points were measured on a Buchi 510
̈
Notes
apparatus. Optical rotations were recorded at 20 °C on a Perkin-Elmer
1
241 polarimeter and are expressed in 10−1 deg cm2 g−1. H and 13C
The authors declare no competing financial interest.
NMR spectra were recorded on a Bruker DPX-200 or Avance-400
spectrometer; chemical shifts (δ) are reported in ppm downfield from
tetramethylsilane (TMS) as internal standard (s singlet, d doublet, t
triplet, q quartet, m multiplet, br broad signal). Two-dimensional
NMR techniques (COSY, HMBC, HSQC) were used to aid in the
ACKNOWLEDGMENTS
■
M.L.W. has been supported by Medical Scientist Training
Program Training Grant, Case Western Reserve University-
T32 GM07250. K.P.W. is funded by the Veterans Affairs Career
Development Program. R.A.B. is funded by Veterans Affairs
Merit Review Program, VISN 10 GRECC, and NIH grants
1R01-A1063517 and 1R01-A100560. The content is solely the
responsibility of the authors and does not necessarily represent
the official views of the National Institutes of Health.
1
assignment of signals in H and 13C spectra. Particularly, in the 13C
spectra, the signal of the boron-bearing carbon atom tend to be
broadened, often beyond the detection limit; however, its resonance
was unambiguously determined by HSQC. Mass spectra were
determined on a gas chromatograph HP 5890 with mass spectrometer
detector HP 5972 (EI, 70 eV) or on an Agilent Technologies LC-
MS(n) Ion Trap 6310A. Elemental analyses were performed on a
Carlo Erba Elemental Analyzer 1110; elemental analyses for the
compounds were within 0.4% of the theoretical values. Elemental
analyses of free boronic acids 3a−d were not obtainable because of the
formation of dehydration products. Nevertheless, these boronic acids
could be converted into analytically pure pinacol/pinanediol esters by
simple exposure to equimolar amount of pinanediol in anhydrous
THF. The starting (+)-pinanediol (1R)-1-(N-bis(trimethylsilyl)-
amino)-2-phenylethaneboronate 1 was obtained as already de-
scribed.49 Full synthesis of all compounds provided in Supporting
Information.
ABBREVIATIONS USED
■
SHV, sulfhydryl-variable; IR, inhibitor resistant; TEM,
Temoneira; MIC, minimum inhibitory concentration; amp,
ampicillin; pip, piperacillin; thin, cephalothin; clav, clavulanate;
sul, sulbactam; tazo, tazobactam; MM/MDS, molecular
modeling/molecular dynamics simulation; BATSI, boronic
acid transition state inhibitor; LB, lysogeny broth; CD, circular
dichroism; MH, Muller−Hinton; SOB, superoptimal broth;
IPTG, isopropyl β-D-1-thiogalactopyranoside; SDS-PAGE,
sodium dodecyl sulfate polyacrylamide gel electrophoresis;
Tm, melting temperature; fu, fraction of denatured protein;
ΔHVH, van’t Hoff enthalpy; ΔSu, entropy of unfolding; ΔΔGu,
Gibbs free energy of unfolding; DS3.1, Discovery Studio 3.1;
THF, tetrahydrofuran; TLC, thin layer chromatography; TMS,
Crystallization, Data Collection, and Refinement. The K234R
protein was crystallized using the vapor diffusion method in 24-well
sitting drop trays (Hampton Research, Aliso Viejo, CA). Well solution
(500 μL) consisted of 21−30% PEG6K and 0.1 M HEPES pH 6.6−
7.6. Five mg/mL K234R was combined with Cymal-6 (Hampton
Research, Aliso Viejo, CA) at a final concentration of 0.56 mM and
1095
dx.doi.org/10.1021/jm301490d | J. Med. Chem. 2013, 56, 1084−1097

Journal of Medicinal Chemistry
Article
Substitutions at Ambler Position 244. Biochemistry 2007, 46, 11361−
11368.
(18) Thomson, J.; Distler, A. M.; Prati, F.; Bonomo, R. A. Probing
Active Site Chemistry in SHV β-Lactamase Variants at Ambler
Position 244. J. Biol. Chem. 2006, 281, 26734−26744.
(19) Sun, T.; Bethel, C. R.; Bonomo, R. A.; Knox, J. R. Inhibitor-
Resistant Class A β-Lactamases: Consequences of the Ser130-to-Gly
Mutation Seen in Apo and Tazobactam Structures of the SHV-1
Variant. Biochemistry 2004, 43, 14111−14117.
(20) Helfand, M. S.; Taracila, M. A.; Totir, M. A.; Bonomo, R. A.;
Buynak, J. D.; van den Akker, F.; Carey, P. R. Raman Crystallographic
Studies of the Intermediates Formed by Ser130Gly SHV, a β-
Lactamase that Confers Resistance to Clinical Inhibitors. Biochemistry
2007, 46, 8689−8699.
(21) Helfand, M. S.; Bethel, C. R.; Hujer, A. M.; Hujer, K. M.;
Anderson, V. E.; Bonomo, R. A. Understanding Resistance to β-
Lactams and β-Lactamase Inhibitors in the SHV β-Lactamase: Lessons
from the Mutagenesis of SER-130. J. Biol. Chem. 2003, 278, 52724−
52729.
(22) Ambler, R. P.; Coulson, A. F. W.; Frere, J.-M.; Ghuysen, J.-M.;
Joris, B.; Forsman, M.; Levesque, R. C.; Tiraby, G.; Waley, S. G. A
Standard Numbering Scheme for the Class A β-lactamases. Biochem. J.
1991, 276, 269−272.
(23) Ghuysen, J.-M. Serine β-Lactamases and Penicillin-Binding
Proteins. Annu. Rev. Microbiol. 1991, 45, 37−67.
(24) Huang, W.; Petrosino, J.; Hirsch, M.; Shenkin, P. S.; Palzkill, T.
Amino acid sequence determinants of β-lactamase structure and
activity. J. Mol. Biol. 1996, 258, 688−703.
(25) Nukaga, M.; Mayama, K.; Hujer, A. M.; Bonomo, R. A.; Knox, J.
R. Ultrahigh Resolution Structure of a Class A β-Lactamase: On the
Mechanism and Specificity of the Extended-Spectrum SHV-2 Enzyme.
J. Mol. Biol. 2003, 328, 289−301.
(26) Thomson, J. M.; Prati, F.; Bethel, C. R.; Bonomo, R. A. Use of
Novel Boronic Acid Transition State Inhibitors to Probe Substrate
Affinity in SHV-Type Extended Spectrum β-Lactamases. Antimicrob.
Agents Chemother. 2007, 51, 1577−1579.
(27) Crompton, I. E.; Cuthbert, B. K.; Lowe, G.; Waley, S. G. The
Inhibition of Serine β-lactamases by Specific Boronic Acids. Biochem. J.
1988, 251, 453−459.
(28) Meroueh, S. O.; Fisher, J. F.; Schlegel, H. B.; Mobashery, S. Ab
Initio QM/MM Study of Class A β-Lactamase Acylation: Dual
Participation of Glu166 and Lys73 in a Concerted Base Promotion of
Ser70. J. Am. Chem. Soc. 2005, 127, 15397−15407.
(29) Wang, X.; Minasov, G.; Shoichet, B. K. The Structural Bases of
Antibiotic Resistance in the Clinically Derived Mutant β-Lactamases
TEM-30, TEM-32, and TEM-34. J. Biol. Chem. 2002, 277, 32149−
32156.
tetramethylsilane; SSRL, Stanford Synchrotron Radiation Light
Source; QM, quantum mechanical
REFERENCES
■
(1) Klevens, R. M.; Edwards, J. R.; Richards, C. L.; Horan, T. C.;
Gaynes, R. P.; Pollock, D. A.; Cardo, D. M. Estimating Health Care-
Associated Infections and Deaths in U.S. Hospitals, 2002. Public
Health Rep. 2007, 122, 160−166.
(2) Drawz, S.; Bonomo, R. Three Decades of β-Lactamase Inhibitors.
Clin. Microbiol. Rev. 2010, 23, 160−201.
(3) Buynak, J. D. Cutting and Stitching: The Cross-linking of
Peptidoglycan in the Assembly of the Bacterial Cell Wall. ACS Chem.
Biol. 2007, 2, 602−605.
(4) Meroueh, S. O.; Bencze, K. Z.; Hesek, D.; Lee, M.; Fisher, J. F.;
Stemmler, T. L.; Mobashery, S. Three-Dmensional Structure of the
Bacterial Cell Wall Peptidoglycan. Proc. Natl. Acad. Sci. U. S. A. 2006,
103, 4404−4409.
(5) Lakaye, B.; Dubus, A.; Lepage, S.; Groslambert, S.; Frere, J. M.
When Drug Inactivation Renders the Target Irrelevant to Antibiotic
Resistance: A Case Story with β-Lactams. Mol. Microbiol. 1999, 31,
89−101.
(6) Lovering, A. L.; Safadi, S. S.; Strynadka, N. C. Structural
Perspective of Peptidoglycan Biosynthesis and Assembly. Annu. Rev.
Biochem. 81, 451−478.
(7) Paterson, D. L.; Bonomo, R. A. Extended-Spectrum β-
Lactamases: A Clinical Update. Clin. Microbiol. Rev. 2005, 18, 657−
686.
(8) Drawz, S.; Bethel, C. R.; Hujer, K. M.; Hurless, K. N.; Distler, A.
M.; Caselli, E.; Prati, F.; Bonomo, R. A. The Role of a Second-Shell
Residue in Modifying Substrate and Inhibitor Interactions in the SHV
β-Lactamase: A Study of Ambler Position Asn276. Biochemistry 2009,
48, 4557−4566.
(9) Perez-Llarena, F. J.; Bou, G. β-Lactamase inhibitors: the story so
far. Curr. Med. Chem. 2009, 16, 3740−3765.
(10) Manageiro, V.; Ferreira, E.; Cougnoux, A.; Albuquerque, L.;
Canica, M.; Bonnet, R. Characterization of the Inhibitor-Resistant
SHV β-Lactamase SHV-107 in a Clinical Klebsiella pneumoniae Strain
Coproducing GES-7 Enzyme. Antimicrob. Agents Chemother. 2012, 56,
1042−1046.
(11) Manageiro, V.; Ferreira, E.; Albuquerque, L.; Bonnet, R.; Canica,
M. Biochemical Study of a New Inhibitor-Resistant β-Lactamase, SHV-
84, Produced by a Clinical Escherichia coli Strain. Antimicrob. Agents
Chemother. 2010, 54, 2271−2272.
(12) Mendonca, N.; Manageiro, V.; Robin, F.; Salgado, M. J.;
Ferreira, E.; Canica, M.; Bonnet, R. The Lys234Arg Substitution in the
Enzyme SHV-72 Is a Determinant for Resistance to Clavulanic Acid
Inhibition. Antimicrob. Agents Chemother. 2008, 52, 1806−1811.
(13) Prinarakis, E. E.; Miriagou, V.; Tzelepi, E.; Gazouli, M.;
Tzouvelekis, L. S. Emergence of an Inhibitor-Resistant β-Lactamase
(SHV-10) Derived from an SHV-5 Variant. Antimicrob. Agents
Chemother. 1997, 41, 838−840.
(14) Dubois, V.; Poirel, L.; Demarthe, F.; Aprin, C.; Coulange, L.;
Minarini, L. A. R.; Bezian, M.-C.; Nordmann, P.; Quentin, C.
Molecular and Biochemical Characterization of SHV-56, a Novel
Inhibitor-Resistant β-Lactamase from Klebsiella pneumoniae. Antimi-
crob. Agents Chemother. 2008, 52, 3792−3794.
(15) Dubois, V.; Poirel, L.; Arpin, C.; Coulange, L.; Bebear, C.;
Nordmann, P.; Quentin, C. SHV-49, a Novel Inhibitor-Resistant β-
Lactamase in a Clinical Isolate of Klebsiella pneumoniae. Antimicrob.
Agents Chemother. 2004, 48, 4466−4469.
(16) Sulton, D.; Pagan-Rodriguez, D.; Zhou, X.; Liu, Y.; Hujer, A. M.;
Bethel, C. R.; Helfand, M. S.; Thomson, J. M.; Anderson, A. E.;
Buynak, J. D.; Ng, L. M.; Bonomo, R. A. Clavulanic Acid Inactivation
of SHV-1 and the Inhibitor-Resistant S130G SHV-1 β-Lactamase. J.
Biol. Chem. 2005, 280, 35528−35536.
(17) Thomson, J.; Distler, A. M.; Bonomo, R. A. Overcoming
Resistance to β-Lactamase Inhibitors: Comparing Sulbactam to Novel
Inhibitors against Clavulanate Resistant SHV Enzymes with
(30) Zafaralla, G.; Manavathu, E. K.; Lerner, S. A.; Mobashery, S.
Elucidation of the Role of Arginine-244 in the Turnover Processes of
Class A β-Lactamases. Biochemistry 1992, 31, 3847−3852.
(31) Hujer, A. M.; Hujer, K. M.; Helfand, M. S.; Anderson, V. E.;
Bonomo, R. A. Amino Acid Substitutions at Ambler Position Gly238
in the SHV-1 β-Lactamase: Exploring Sequence Requirements for
Resistance to Penicillins and Cephalosporins. Antimicrob. Agents
Chemother. 2002, 46, 3971−3977.
(32) Hujer, A. M.; Bethel, C. R.; Bonomo, R. A. Antibody Mapping
of the Linear Epitopes of CMY-2 and SHV-1 β-Lactamases.
Antimicrob. Agents Chemother. 2004, 48, 3980−3988.
(33) Papp-Wallace, K. M.; Taracila, M. A.; Smith, K. M.; Xu, Y.;
Bonomo, R. A. Understanding the Molecular Determinants of
Substrate and Inhibitor Specificities in the Carbapenemase KPC-2:
Exploring the Roles of Arg220 and Glu276. Antimicrob. Agents
Chemother. 2012, 56, 4428−4438.
(34) Methods for Dilution Antimicrobial Susceptibility Tests for Bacteria
That Grow Aerobically, 1st ed.; Approved Standard: Clinical and
Laboratory Standards Institute: Wayne, PA, 2008.
(35) Lin, S.; Thomas, M.; Shlaes, D. M.; Rudin, S. D.; Knox, J. R.;
Anderson, V.; Bonomo, R. A. Kinetic Analysis of an Inhibitor-Resistant
1096
dx.doi.org/10.1021/jm301490d | J. Med. Chem. 2013, 56, 1084−1097

Journal of Medicinal Chemistry
Article
Variant of the OHIO-1 β-Lactamase, An SHV-Family Class A Enzyme.
Biochem. J. 1998, 333, 395−400.
(36) Morandi, F.; Caselli, E.; Morandi, S.; Focia, P. J.; Blazquez, J.;
Shoichet, B. K.; Prati, F. Nanomolar Inhibitors of AmpC β-Lactamase.
J. Am. Chem. Soc. 2003, 125, 685−695.
(37) De Meester, F.; Joris, B.; Reckinger, G.; Bellefroid-Bourguignon,
C.; Frere, J.-M.; Waley, S. G. Automated Analysis of Enzyme
Inactivation Phenomena: Application to β-Lactamases and DD-
Peptidases. Biochem. Pharmacol. 1987, 36, 2393−2403.
(38) Papp-Wallace, K. M.; Taracila, M.; Wallace, C. J.; Hujer, K. M.;
Bethel, C. R.; Hornick, J. M.; Bonomo, R. Elucidating the role of
Trp105 in the KPC-2 β-lactamase. Protein Sci. 2010, 19, 1714−1727.
(39) Papp-Wallace, K. M.; Taracila, M.; Hornick, J. M.; Hujer, A. M.;
Hujer, K. M.; Distler, A. M.; Endimiani, A.; Bonomo, R. A. Substrate
Selectivity and a Novel Role in Inhibitor Discrimination by Residue
237 in the KPC-2 β-Lactamase. Antimicrob. Agents Chemother. 2010,
54, 2867−2877.
(40) Nagarajan, R.; Pratt, R. F. Thermodynamic Evaluation of a
Covalently Bonded Transition State Analogue Inhibitor: Inhibition of
β-Lactamases by Phosphonates. Biochemistry 2004, 43, 9664−9673.
(41) Becktel, W.; Schellman, J. Protein Stability Curves. Biopolymers
1987, 26, 1859−1877.
(42) Beadle, B. M.; McGovern, S. L.; Patera, A.; Shoichet, B. K.
Functional Analyses of AmpC β-Lactamase through Differential
Stability. Protein Sci. 1999, 8, 1816−1824.
(43) Papp-Wallace, K. M.; Bethel, C. R.; Distler, A. M.; Kasuboski,
C.; Taracila, M.; Bonomo, R. A. Inhibitor Resistance in the KPC-2 β-
Lactamase, a Preeminent Property of This Class A β-Lactamase.
Antimicrob. Agents Chemother. 2010, 54, 890−897.
(44) Wu, G.; Robertson, D. H.; Brooks, C. L. R.; Vieth, M. Detailed
Analysis of Grid-Based Molecular Docking: A Case Study of
CDOCKER-A CHARMm-Based MD Docking Algorithm. J. Comput.
Chem. 2003, 24, 1549−1562.
(45) Drawz, S. M.; Taracila, M.; Caselli, E.; Prati, F.; Bonomo, R. A.
Exploring Sequence Requirements for C3/C4 Carboxylate Recog-
nition in the Pseudomonas aeruginosa Cephalosporinase: Insights into
Plasticity of the AmpC β-Lactamase. Protein Sci. 2011, 20, 941−958.
(46) Verlet, L. Computer “Experiments” on Classical Fluids. I.
Thermodynamical Properties of Lennard-Jones Molecules. Phys. Rev.
1967, 159, 98−103.
(47) Allen, M. P.; Tildesley, D. J. Periodic Boundary Conditions and
Potential Truncation. In Computer Simulation of Liquids; Oxford
University Press: Oxford, UK, 1987.
(48) Haile, J. M. Molecular Dynamic Simulations: Elementary Methods;
John Wiley & Sons: New York, 1992.
(49) Eidam, O.; Romagnoli, C.; Caselli, E.; Babaoglu, K.; Teotico
Pohlhaus, D.; Karpiak, J.; Bonnet, R.; Shoichet, B. K.; Prati, F. Design,
Synthesis, Crystal Structures, and Antimicrobial Activity of Sulfona-
mide Boronic Acids as β-Lactamase Inhibitors. J. Med. Chem. 2010, 53,
7852−7863.
(50) Otwinowski, Z.; Minor, W. [20] Processing of X-ray Diffraction
Data Collected in Oscillation Mode. Methods Enzymol. 1997, 276,
307−326.
(51) Winn, M. D.; Ballard, C. C.; Cowtan, K. D.; Dodson, E. J.;
Emsley, P.; Evans, P. R.; Keegan, R. M.; Krissinel, E. B.; Leslie, A. G.;
McCoy, A.; McNicholas, S. J.; Murshudov, G. N.; Pannu, N. S.;
Potterton, E. A.; Powell, H. R.; Read, R. J.; Vagin, A.; Wilson, K. S.
Overview of the CCP4 Suite and Current Developments. Acta
Crystallogr., Sect. D: Biol. Crystallogr. 2011, 67, 235−242.
(52) Emsley, P.; Cowtan, K. Coot: Model-Building Tools for
Molecular Graphics. Acta Crystallogr., Sect. D: Biol. Crystallogr. 2004,
60, 2126−2132.
(53) Shuttelkopf, A. W.; van Aalten, D. M. PRODRG: A Tool for
High-Throughput Crystallography of Protein−Ligand Complexes.
Acta Crystallogr., Sect. D: Biol. Crystallogr. 2004, 60, 1355−1363.
1097
dx.doi.org/10.1021/jm301490d | J. Med. Chem. 2013, 56, 1084−1097